Opus 5 max thinking reportedly underperforms xhigh on 20–30% of benchmarks
keunwoochoi · x · 2026-07-25
A reply to a discussion of Opus 5 notes that Opus 4.8 does not show the same issue and suggests the behavior may indicate a rushed release.
The quoted observation underneath says that in roughly 20–30% of reported benchmarks, Opus 5 max thinking performs worse than xhigh. The surprising part is that more thinking and more test-time compute would normally be expected to improve results, so the benchmark drops may point to either an emergent issue in smaller models or a post-training problem.
Related event: Opus 5 Coding Paradox: Higher Reasoning Leads to Lower Scores(13 posts)→
More from Models
- Astra Scores 83% on GauntletBench, First Computer-Use Agent to Beat Human Baseline — ducha_aiki · 2026-09-11
- Kimi K2.8 Preview rolls out: near-K3 coding performance, 1M context for all tiers — teortaxesTex · 2026-09-11
- Looking for a classifier of software engineering task shapes to pick models per task — StewartalsopIII · 2026-09-11
- DeepSeek V4 Pro API to continue after Sept 2026, billing unchanged — teortaxesTex · 2026-09-11
- DeepSeek V4.1 Flash Hits 98% of GPT-6 Astra's Score at 1.4% of the Cost in Third-Party Benchmark — ayushtweetshere · 2026-09-11
- TheZvi Polls: Has Your Coding Model Choice Changed Since Fable 5.1 and Astra? — TheZvi · 2026-09-11